Loss Design and Architecture Selection for Long-Tailed Multi-Label Chest X-Ray Classification

Nikhileswara Rao Sulake

eess.IVMar 2, 2026

Loss Design and Architecture Selection for Long-Tailed Multi-Label Chest X-Ray Classification

Authors: Nikhileswara Rao Sulake

Organizations: Department of Computer Science and Engineering, Rajiv Gandhi University of Knowledge Technologies, Nuzvid, India

Abstract

Long-tailed class distributions pose a significant challenge for multi-label chest X-ray (CXR) classification, where rare but clinically important findings are severely underrepresented. In this work, we present a systematic empirical evaluation of loss functions, CNN backbone architectures and post-training strategies on the CXR-LT 2026 benchmark, comprising approximately 143K images with 30 disease labels from PadChest. Our experiments demonstrate that LDAM with deferred re-weighting (LDAM-DRW) consistently outperforms standard BCE and asymmetric losses for rare class recognition. Amongst the architectures evaluated, ConvNeXt-Large achieves the best single-model performance with 0.5220 mAP and 0.3765 F1 on our development set, whilst classifier re-training and test-time augmentation further improve ranking metrics. On the official test leaderboard, our submission achieved 0.3950 mAP, ranking 5th amongst all 68 participating teams with total of 1528 submissions. We provide a candid analysis of the development-to-test performance gap and discuss practical insights for handling class imbalance in clinical imaging settings. Code is available at https://github.com/Nikhil-Rao20/Long_Tail.

Figures & tables

Model	Loss	Post-training	AP	AUC	F1	ECE
ResNet-50	BCE	–	$0.3248$	$0.8410$	$0.3222$	$0.8884$
ResNet-50	Asymmetric	–	$0.0667$	$0.5603$	$0.0843$	$0.9526$
ResNet-50	LDAM+DRW	–	$0.4241$	$0.8435$	$0.2676$	$0.5575$
ResNet-50	LDAM+DRW	cRT	$0.4303$	$0.8828$	$0.3233$	$0.8300$
ResNet-50	LDAM+DRW	cRT + TTA	$0.4325$	$0.8864$	$0.3102$	$0.8247$
ResNet-101	LDAM+DRW	–	$0.4584$	$0.8679$	$0.2564$	$0.5332$

Table 1: Development set performance comparison across architectures, loss functions, and post-training strategies. Best AP and F1 are bolded.

Figure 1: Class-activation maps overlaid on CXR test images (predicted label and probability shown). The model localizes many findings correctly but probability calibration and thresholding cause instance-level misses. Rows correspond to different findings (kyphosis, hernia, azygos lobe).

Rank	Team	Affiliation	mAP $\uparrow$	AUC $\uparrow$	F1 $\uparrow$
1	CVMAIL x MIHL	Vietnam National University, Vietnam	0.5854	0.9259	0.3518
2	Cool Peace	KAIST Graduate School of AI, Korea	0.4827	0.9186	0.3162
3	VIU	Vietnam National University, Vietnam	0.4599	0.8827	0.4504
4	Bibimbap-Bueno	Case Western Reserve University, USA	0.4297	0.8753	0.2482
5	Nikhil Rao Sulake	RGUKT Nuzvid, India	0.3950	0.8591	0.0945
6	UGIVIA team	Universitat de les Illes Balears, Spain	0.2362	0.7756	0.2353

Table 2: Official CXR-LT 2026 Test Leaderboard (Task 1: In-distribution Multi-label Classification). Primary ranking metric: macro-averaged mAP.

Explore similar work

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification
Apr 16, 2026Hexin Dong, Yi Lin, Pengyu Zhou +25Zero-Shot Learning Long-Tailed Classification
Department of Population Health Sciences, Weill Cornell Medicine, New York, USA · Department of Radiology, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China · Department of Radiology, West China School of Medicine, Sichuan University, Sichuan University Affiliated Chengdu Second People’s Hospital, Chengdu Second People’s Hospital, Chengdu, China +20
TRCGL-Net: A Long-Tailed Multi-Label Chest X-Ray Classification Framework with Generative Data Augmentation and Label Co-Occurrence Modeling
Jul 1, 2026Tong Shao, Hongshun Ling, Li Zhang +4Long-Tailed Classification Chest X-Ray Classification
School of Biomedical Engineering, South-Central Minzu University, Wuhan 430074, China
Overview of the CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification
Feb 25, 2026Hexin Dong, Yi Lin, Pengyu Zhou +7Chest X-Ray Classification Medical Image Benchmarks
Weill Cornell Medicine · Peking Union Medical College · University of Minnesota +2

Apr 16, 2026J/K move · Enter open · S save

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification

Hexin Dong, Yi Lin, Pengyu Zhou, Fengnian Zhao+24

Department of Population Health Sciences, Weill Cornell Medicine, New York, USA · Department of Radiology, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China · Department of Radiology, West China School of Medicine, Sichuan University, Sichuan University Affiliated Chengdu Second People’s Hospital, Chengdu Second People’s Hospital, Chengdu, China+20

Zero-Shot Learning Long-Tailed Classification Chest X-Ray Classification Domain Generalization Open-Set Recognition Medical Image Benchmarks

Open paper AskPDF arXiv

Chest X-ray (CXR) interpretation is hindered by the long-tailed distribution of pathologies and the open-world nature of clinical environments. Existing benchmarks often rely on closed-set classes from a single institution, failing to capture the prevalence of rare diseases or the appearance of novel findings. To address this, we present the CXR-LT challenge. The first event, CXR-LT 2023, established a large-scale benchmark for long-tailed multi-label CXR classification and identified key challenges in rare disease recognition. CXR-LT 2024 further expanded the label space and introduced a zero-shot task to study generalization to unseen findings. Building on the success of CXR-LT 2023 and 2024, this third iteration of the benchmark introduces a multi-center dataset comprising over 145,000 images from PadChest and NIH Chest X-ray datasets. Additionally, all development and test sets in CXR-LT 2026 are annotated by radiologists, providing a more reliable and clinically grounded evaluation than report-derived labels. The challenge defines two core tasks this year: (1) Robust Multi-Label Classification on 30 known classes and (2) Open-World Generalization to 6 unseen (out-of-distribution) rare disease classes. This paper summarizes the overview of the CXR-LT 2026 challenge. We describe the data collection and annotation procedures, analyze solution strategies adopted by participating teams, and evaluate head-versus-tail performance, calibration, and cross-center generalization gaps. Our results show that vision-language foundation models improve both in-distribution and zero-shot performance, but detecting rare findings under multi-center shift remains challenging. Our study provides a foundation for developing and evaluating AI systems in realistic long-tailed and open-world clinical conditions.

Abstract

Explore similar work

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification

TRCGL-Net: A Long-Tailed Multi-Label Chest X-Ray Classification Framework with Generative Data Augmentation and Label Co-Occurrence Modeling

Overview of the CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification